| Time |
Nick |
Message |
| 01:04 |
|
proller joined #minetest-dev |
| 01:30 |
|
SFENCE joined #minetest-dev |
| 02:17 |
|
SFENCE joined #minetest-dev |
| 02:25 |
|
ivanbu joined #minetest-dev |
| 02:25 |
|
MTDiscord1 joined #minetest-dev |
| 04:19 |
|
v-rob joined #minetest-dev |
| 05:00 |
|
MTDiscord joined #minetest-dev |
| 05:00 |
|
TheCoffeMaker_ joined #minetest-dev |
| 05:49 |
|
SFENCE joined #minetest-dev |
| 06:51 |
|
SFENCE joined #minetest-dev |
| 07:01 |
|
SFENCE joined #minetest-dev |
| 07:05 |
|
SFENCE joined #minetest-dev |
| 07:07 |
|
YuGiOhJCJ joined #minetest-dev |
| 07:11 |
|
SFENCE joined #minetest-dev |
| 07:12 |
|
SFENCE joined #minetest-dev |
| 07:27 |
|
SFENCE joined #minetest-dev |
| 07:34 |
|
calcul0n joined #minetest-dev |
| 10:45 |
|
SFENCE joined #minetest-dev |
| 13:33 |
|
proller joined #minetest-dev |
| 13:47 |
|
Lupercus joined #minetest-dev |
| 14:17 |
|
SFENCE joined #minetest-dev |
| 15:11 |
|
appguru joined #minetest-dev |
| 15:27 |
|
TheCoffeMaker joined #minetest-dev |
| 16:20 |
|
SFENCE joined #minetest-dev |
| 16:31 |
|
SFENCE joined #minetest-dev |
| 16:42 |
|
SFENCE joined #minetest-dev |
| 16:45 |
|
SFENCE joined #minetest-dev |
| 17:09 |
|
Desour joined #minetest-dev |
| 17:11 |
|
SFENCE joined #minetest-dev |
| 18:07 |
ROllerozxa |
is there a preprocessor define for checking if minetest is being built with SDL2 or not? |
| 18:10 |
ROllerozxa |
I know _IRR_COMPILE_WITH_SDL_DEVICE_ exists in irrlichtmt but would that be dirty to introduce into minetest |
| 18:12 |
Krock |
RenderingEngine::get_raw_device()->getType() == EIDT_SDL for the active render. I cannot find any check for support |
| 18:13 |
Krock |
you could probe it by calling createDeviceEx using SDL ... |
| 18:14 |
Krock |
* params.DeviceType == EIDT_SDL |
| 18:14 |
ROllerozxa |
preprocessor define. I can't wrap a header include within that without causing errors |
| 18:15 |
ROllerozxa |
(this is for fixing building minetest w/ SDL2 on MSYS2, it requires including the SDL.h header in minetest main.cpp for SDL2main to work) |
| 18:15 |
ROllerozxa |
would adding an USE_SDL2 to cmake_config.h.in be ok |
| 18:15 |
Krock |
is there no way to implement it in Irrlicht? |
| 18:16 |
Krock |
because Minetest currently does not (and should not) care which device is used |
| 18:17 |
rubenwardy |
Minetest will need to care when it starts using SDL directly |
| 18:17 |
rubenwardy |
ie: for gamepad input |
| 18:17 |
ROllerozxa |
alternative would be https://github.com/minetest/irrlicht/pull/265 which seems to be the wrong way to do it |
| 18:18 |
rubenwardy |
for my SDL gamepad PR, I just break compiling on non-SDL |
| 18:18 |
Krock |
it might be possible to forward the define by using the IrrConfig something header |
| 18:19 |
Krock |
IrrCompileConfig.h |
| 18:19 |
Krock |
no that's not it. |
| 18:23 |
sfan5 |
do we need to use sdl2main? iirc it was optional |
| 18:23 |
Krock |
either way, the irrlicht PR does look correct. After all, SDL2::SDL2 is supposed to provide necessary include and lirary paths |
| 18:24 |
|
v-rob joined #minetest-dev |
| 18:24 |
Krock |
> Under Visual C++, you need to link with SDL2main.lib |
| 18:24 |
Krock |
https://wiki.libsdl.org/SDL2/FAQWindows |
| 18:24 |
Krock |
however, this is about msys2 so that does not apply. |
| 18:25 |
Krock |
unless they mean the VC++ runtime and not MSVC |
| 18:27 |
sfan5 |
optional in the sense that it is then your responsibility to adhere to platform specific init procedures |
| 18:27 |
sfan5 |
> SDL2main provides SDL_main(), which is a global entry point for all SDL apps. You are not required/forced to use it, but is presence is based on the variety of the systems SDL supports. Windows uses WinMain(), Linux uses main(), Android needs JNI and some Java to actually use SDL, so SDL_main() makes things a little easier. |
| 18:27 |
sfan5 |
(2013 forum post) |
| 18:33 |
ROllerozxa |
the only other platform that has init stuff in SDL2main that we would reasonably support other than windows would be haiku |
| 18:35 |
|
SFENCE joined #minetest-dev |
| 18:36 |
v-rob |
> Minetest will need to care when it starts using SDL directly |
| 18:36 |
v-rob |
Yes, and my events code for the formspec replacement (which isn't in the current PR) uses SDL_Event heavily. I want to circumvent Irrlicht's event system from the get-go. |
| 18:37 |
ROllerozxa |
I guess we could also stall and wait until SDL3 comes out which won't have this SDL2main library, sounds like the most minetesty option |
| 18:37 |
v-rob |
After all, the new HUD/GUI code is the whole reason I started pushing for SDL from the beginning |
| 18:37 |
rubenwardy |
let's not wait for v3 for projects which have an association with Valve |
| 18:41 |
v-rob |
And SDL means that I can integrate #8679 into the UI API with both keycodes/scancodes. Woo! |
| 18:41 |
ShadowBot |
https://github.com/minetest/minetest/issues/8679 -- Add FormSpec key and mouse events by v-rob |
| 18:41 |
v-rob |
Yes, I am very happy about using SDL directly. |
| 18:47 |
sfan5 |
btw (I have also commented this somewhere) I have built mt + sdl2 fine with mingw no changes to the current build script |
| 18:47 |
sfan5 |
we can also take sdl2main out if that doesn't break anything else |
| 18:51 |
ROllerozxa |
speaking about breaking things, I reopened https://github.com/minetest/irrlicht/pull/265 and the linux SDL CI now fails after I resolved merge conflicts, don't know why that happened |
| 19:01 |
Krock |
possibly FindSDL2.cmake does not support the SDL2::SDL2 target yet in the CMake version that is used in the buildbot |
| 19:02 |
Krock |
"-- Found SDL2: -lSDL2" is unlike the other outputs |
| 19:23 |
|
v-rob joined #minetest-dev |
| 19:36 |
|
TheCoffeMaker joined #minetest-dev |
| 19:43 |
|
TheCoffeMaker joined #minetest-dev |
| 20:33 |
|
SFENCE joined #minetest-dev |
| 22:18 |
|
SFENCE joined #minetest-dev |
| 23:26 |
|
YuGiOhJCJ joined #minetest-dev |
| 23:32 |
|
panwolfram joined #minetest-dev |
